【摘要】 针对基本粒子群优化(PSO)算法在解决复杂多峰问题时易于陷入局部最优解的问题,提出一种基于爆炸冲击波模型的PSO算法(简称BW-PSO算法)。该算法通过加入种群多样性监督条件,使得当种群数量缩小至给定阈值时,触发粒子冲击波过程:最优粒子与次优粒子进行交叉变异,处于爆炸半径内的粒子受到牵引力,加速收敛至当前极值;处于爆炸半径外的粒子受到冲击力向外扩散,增加了找到全局最优值的可能性。BW-PSO算法不仅能够通过最优粒子变异操作提升当前解的精度,而且通过粒子冲击波过程,增加了种群多样性,提升了粒子对全局空间开发的能力。实验结果表明,基于爆炸冲击波模型的PSO算法在求解多峰问题表现优于变异PSO算法与带电PSO算法。更多还原

【Abstract】 A new Particle Swarm Optimization( PSO) algorithm based on the blast wave model( referred to as BW-PSO algorithm) was proposed aiming at the problem that the basic PSO algorithm when solving complex multimodal problems is easy to fall into local optimal solution. The supervision conditions of population diversity were added to the basic PSO algorithm so that the process of particle shock was triggered when the population decreased to a given threshold value. Crossover and mutation occurred between optimal and suboptimal particles so that the particles within the blast radius by the traction were subjected to accelerate convergence to the current extreme and the particles outside the blast radius were subjected to spread out, which increased the possibility of finding the global optimum value. BW-PSO algorithm not only improved the accuracy of the current solution by the mutation between optimal and suboptimal particles, but also increased the population diversity with the shock wave process of the particles and enhances the ability of the global space development of the particles. Compared with the mutative PSO and charged PSO, the results indicate that the BW-PSO algorithm has a better performance to solve multi-modal optimization problem.更多还原